def factor_number(m, true_p, true_q, use_true_values=False):
apply_preprocessing = True
preprocessing_verbose = False
optimization_verbose = False
if m == 56153:
p_dict, q_dict, z_dict, clauses = factor_56153()
elif m == 291311:
p_dict, q_dict, z_dict, clauses = factor_291311()
elif use_true_values:
p_dict, q_dict, z_dict, clauses = create_clauses(m, true_p, true_q, apply_preprocessing, preprocessing_verbose)
else:
p_dict, q_dict, z_dict, clauses = create_clauses(m, None, None, apply_preprocessing, preprocessing_verbose)
number_of_uknowns, number_of_carry_bits = calculate_number_of_unknowns(p_dict, q_dict, z_dict)
print("Number of unknowns:", number_of_uknowns)
print("Number of carry bits:", number_of_carry_bits)
if clauses[0] == 0 and len(set(clauses)) == 1:
if number_of_uknowns == 0:
p, q = decode_solution(p_dict, q_dict)
return p, q, None
optimization_engine = OptimizationEngine(clauses, m, steps=1, grid_size=5, verbose=optimization_verbose, visualize=True)
sampling_results, mapping = optimization_engine.perform_qaoa()
most_frequent_bit_string = max(sampling_results, key=lambda x: sampling_results[x])
squared_overlap = calculate_squared_overlap(mapping, sampling_results, true_p, true_q, p_dict, q_dict)
p_dict = update_dictionary(most_frequent_bit_string, mapping, p_dict)
q_dict = update_dictionary(most_frequent_bit_string, mapping, q_dict)
z_dict = update_dictionary(most_frequent_bit_string, mapping, z_dict)
p, q = decode_solution(p_dict, q_dict)
return p, q, squared_overlap
def main():
threshold = 1e5
primes = get_primes_lower_than_n(int(np.sqrt(threshold)))
primes = primes[1:]
qubits_required_no_preprocessing = []
qubits_required_with_preprocessing = []
initial_time = time.time()
# file_name = "preprocessing_full_results.csv"
# plot_name = "reprocessing_full_plot.png"
file_name = "preprocessing_no_z2_results.csv"
plot_name = "reprocessing_no_z2_plot.png"
for p in primes:
for q in primes:
if p < q:
continue
m = p * q
if m > threshold:
continue
start_time = time.time()
# p_dict, q_dict, z_dict, _ = create_clauses(m, p, q, apply_preprocessing=False, verbose=False)
# x, z = assess_number_of_unknowns(p_dict, q_dict, z_dict)
# qubits_required_no_preprocessing.append([m, x, z])
p_dict, q_dict, z_dict, _ = create_clauses(m, p, q, apply_preprocessing=True, verbose=False)
x, z = assess_number_of_unknowns(p_dict, q_dict, z_dict)
qubits_required_with_preprocessing.append([m, x, z])
end_time = time.time()
t = np.round(end_time - start_time, 3)
print(p, q, m, x, z, t, " ")#, end="\r")
np.savetxt(file_name, np.array(qubits_required_with_preprocessing), delimiter=",", fmt='%.d', header='m,unknowns,carry_bits', comments='')
qubits_required_no_preprocessing = np.genfromtxt('no_preprocessing', skip_header=1, delimiter=',')
# qubits_required_with_preprocessing = np.genfromtxt('preprocessing_no_z2_results', skip_header=1, delimiter=',')
print("Total time:", np.round((end_time - initial_time) / 60, 3), '[min]')
data_1 = np.array(qubits_required_no_preprocessing)
data_2 = np.array(qubits_required_with_preprocessing)
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1)
ax.scatter(data_1[:, 0], data_1[:, 1], label="No classical preprocessing", s=10)
ax.scatter(data_2[:, 0], data_2[:, 1], label="Classical preprocessing", s=10)
ax.set_xlabel("Biprime to be factored")
ax.set_ylabel("Number of qubit required")
ax.set_xscale('log')
plt.legend()
plt.savefig(plot_name)
plt.show()